公开(公告)号：US10268232B2

公开(公告)日：2019-04-23

申请号：US15612043

申请日：2017-06-02

Applicant: Massachusetts Institute of Technology

Inventor： Nicholas Christopher Harris ,  Jacques Johannes Carolan ,  Mihika Prabhu ,  Dirk Robert Englund ,  Scott A. Skirlo ,  Yichen Shen ,  Marin Soljacic

IPC: G02F1/21 ,  G02F3/02 ,  G06E3/00 ,  G06N3/08 ,  G02F1/225 ,  G02F1/365 ,  G06N3/067 ,  G02F1/35

Abstract: An optical neural network is constructed based on photonic integrated circuits to perform neuromorphic computing. In the optical neural network, matrix multiplication is implemented using one or more optical interference units, which can apply an arbitrary weighting matrix multiplication to an array of input optical signals. Nonlinear activation is realized by an optical nonlinearity unit, which can be based on nonlinear optical effects, such as saturable absorption. These calculations are implemented optically, thereby resulting in high calculation speeds and low power consumption in the optical neural network.

公开(公告)号：US10768659B2

公开(公告)日：2020-09-08

申请号：US16273257

申请日：2019-02-12

Applicant: Massachusetts Institute of Technology

Inventor： Jacques Johannes Carolan ,  Mihika Prabhu ,  Scott A. Skirlo ,  Yichen Shen ,  Marin Soljacic ,  Nicholas Christopher Harris ,  Dirk Englund

IPC: G06E3/00 ,  G06N3/04 ,  G06N3/08 ,  G02F1/225 ,  G02F1/35 ,  G02F1/365 ,  G02F3/02 ,  G06N3/067 ,  G02F1/21

Abstract: An optical neural network is constructed based on photonic integrated circuits to perform neuromorphic computing. In the optical neural network, matrix multiplication is implemented using one or more optical interference units, which can apply an arbitrary weighting matrix multiplication to an array of input optical signals. Nonlinear activation is realized by an optical nonlinearity unit, which can be based on nonlinear optical effects, such as saturable absorption. These calculations are implemented optically, thereby resulting in high calculation speeds and low power consumption in the optical neural network.

公开(公告)号：US11334107B2

公开(公告)日：2022-05-17

申请号：US16986383

申请日：2020-08-06

Applicant: Massachusetts Institute of Technology

Inventor： Jacques Johannes Carolan ,  Mihika Prabhu ,  Scott A. Skirlo ,  Yichen Shen ,  Marin Soljacic ,  Dirk Englund ,  Nicholas Christopher Harris

IPC: G06E3/00 ,  G06N3/04 ,  G06N3/08 ,  G02F1/225 ,  G02F1/35 ,  G02F1/365 ,  G02F3/02 ,  G06N3/067 ,  G02F1/21

Abstract: An optical neural network is constructed based on photonic integrated circuits to perform neuromorphic computing. In the optical neural network, matrix multiplication is implemented using one or more optical interference units, which can apply an arbitrary weighting matrix multiplication to an array of input optical signals. Nonlinear activation is realized by an optical nonlinearity unit, which can be based on nonlinear optical effects, such as saturable absorption. These calculations are implemented optically, thereby resulting in high calculation speeds and low power consumption in the optical neural network.

公开(公告)号：US20190294199A1

公开(公告)日：2019-09-26

申请号：US16273257

申请日：2019-02-12

Applicant: Massachusetts Institute of Technology

Inventor： Jacques Johannes Carolan ,  Mihika Prabhu ,  Scott A. Skirlo ,  Yichen Shen ,  Marin Soljacic ,  Nicholas Christopher Harris ,  Dirk Englund

IPC: G06E3/00 ,  G02F3/02 ,  G06N3/067 ,  G02F1/225 ,  G06N3/08 ,  G02F1/35 ,  G02F1/365

Abstract: An optical neural network is constructed based on photonic integrated circuits to perform neuromorphic computing. In the optical neural network, matrix multiplication is implemented using one or more optical interference units, which can apply an arbitrary weighting matrix multiplication to an array of input optical signals. Nonlinear activation is realized by an optical nonlinearity unit, which can be based on nonlinear optical effects, such as saturable absorption. These calculations are implemented optically, thereby resulting in high calculation speeds and low power consumption in the optical neural network.